chris@umcp-cs.UUCP (Chris Torek) (09/09/86)
In article <3945@ut-ngp.UUCP> werner@ut-ngp.UUCP (Werner Uhrig) writes: >My argument is that, as long as there is no second site SEISMO on the net >it is an atrocity for people to get slapped in the face by the mailer which >all of a sudden disavows knowing SEISMO. In the domain scheme of things, whether there is one seismo or not is irrelevant. Those designing domains decided that this was the way to go. When---not if; there is a Final Authority enforcing the changeover---when the ARPA/MILnet Internet converts completely, there WILL BE NO HOST TABLE. It will no longer be possible to tell whether seismo is unique in the first place. >And, BTW, what is your image of how the *average* user finds out >that it is now supposed to be SEISMO.CSS.GOV? For that matter, how does the average user find out about seismo in the first place? (I cannot answer either question.) This problem is really the same one: it has just been repeated. I grant that this *is* annoying. >imagine the terror of *ALL* sites dropping their basic alias This WILL HAPPEN. (Sorry.) In this new scheme of things, if I understand aright, you who are on the host that will be called `ngp.utexas.edu' will be able to refer to other `.utexas.edu' sites, e.g., the one now known as ut-sally, using what you might consider your basic alias. Simply saying `send mail to jsq@sally' will work for you. I, in the land of `.umd.edu', will have to use `mail jsq@sally.utexas.edu'---which address you too may use. Indeed, this `fully qualified' name will work from anywhere in the Internet: That is its very reason for existence. Why the change, if the current system is working so well? The answer is that it is *not* working so well. The pain is well hidden. But it is there, at the NIC, where an enourmous and fast-growing host table must be maintained in a consistent fashion, and must be distributed to thousands of machines. >And dealing with the complexity of multiple aliases in a user-friendly >fashion isn't such a difficult problem either (that is ancient >problem with several acceptable solutions). First, if there are >only few aliases, simple list me a menu of choices to pick from. >Second, if there are lots of "matches" try to narrow down the "hits" >by asking for additional info, possibly listing some classification >keywords. Sound familiar? Probably not to those who have not seen Tenex-style completion. But whether or not it sounds familiar, note that both solutions are predicated upon a centralised information base. One cannot list all possible choices unless one knows all possible choices. In our case (host names), this centralised database does now exist; but it is going away: its doom was deemed long ago. Soon there will be one small centralised database, and many small distributed databases, each describing one domain. Not all will be up at once. Indeed, given the size and reliability of the Internet, it is likely that at least one will be down---unreachable---at any given time. >As the number of users and sites increases we will, of course, approach a >problem not dissimilar to that of sending UPS-mail today. >I'll worry about that when we get there, but nothing I have said is any less >valid in result .... Alas for you, you may begin worrying now. Domains are moving in, and postal mail is the model. -- In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (+1 301 454 1516) UUCP: seismo!umcp-cs!chris CSNet: chris@umcp-cs ARPA: chris@mimsy.umd.edu
johnsson@decwrl.DEC.COM (Richard Johnsson) (09/09/86)
Chris Torek writes (among other things): > In our case (host names), this centralised database does now exist; > but it is going away: its doom was deemed long ago. As I'm sure Chris knows, that centralized database is incomplete and growing more so continuously. I have over 12,000 host names in the DEC.COM domain that are not part of the centralized data, yet they are reachable by hosts using the domain system (via MX records). I'll bet that I have duplicates of many names you thought were unique. (I don't have a SEISMO, however :-) As a matter of fact, the DEC network powers-that-be have decreed that the list of hostnames is proprietary and I have been specifically told I cannot make them public. (Don't flame me, I just work here.) I've been planning to drop the unqualified names from our host table entries for some time. As has been previously been pointed out, they have been obsolete for a long time. -- Richard Johnsson, DEC Western Software Lab, Palo Alto, CA UUCP: {decvax,ucbvax}!decwrl!johnsson ARPA: johnsson@decwrl.dec.com DEC ENet: rhea::johnsson phone: +1 415 853 6676
jsq@im4u.UUCP (John Quarterman) (09/15/86)
In article <3340@umcp-cs.UUCP> chris@umcp-cs.UUCP (Chris Torek) writes: >In this new scheme of things, if I understand aright, you who are >on the host that will be called `ngp.utexas.edu' will be able to >refer to other `.utexas.edu' sites, e.g., the one now known as >ut-sally, using what you might consider your basic alias. Simply >saying `send mail to jsq@sally' will work for you. I, in the land >of `.umd.edu', will have to use `mail jsq@sally.utexas.edu'---which >address you too may use. Indeed, this `fully qualified' name will >work from anywhere in the Internet: That is its very reason for >existence. They've been ngp.utexas.edu and sally.utexas.edu for many months: ut-sally is just sally's UUCP name. Abbreviations are supported locally. Incidentally, I no longer work for UTCS nor am I currently the domain or network administrator there. Some of you may find this of interest: %A John S. Quarterman %A Josiah C. Hoskins, %T Notable Computer Networks %J Communications of the ACM %I Association for Computing Machinery %V 29 %N 10 %P 45pp. %C New York, NY %D October 1986 -- John Quarterman, UUCP: {gatech,harvard,ihnp4,pyramid,seismo}!ut-sally!im4u!jsq ARPA Internet and CSNET: jsq@im4u.UTEXAS.EDU, jsq@sally.UTEXAS.EDU